Retrospective Counterfactual Prediction by Conditioning on the Factual Outcome: A Cross-World Approach
Not on the map. Why?
Abstract (arXiv)
Retrospective causal questions ask what would have happened to an observed individual had they received a different treatment. We study the problem of estimating μ(x,y)=mathbbE[Y(1)mid X=x,Y(0)=y], the expected counterfactual outcome for an individual with covariates x and observed outcome y, and constructing valid prediction intervals under the Neyman-Rubin superpopulation model. This quantity is generally not identified without additional assumptions. To link the observed and unobserved potential outcomes, we work with a cross-world correlation ρ(x)=cor(Y(1),Y(0)mid X=x); plausible bounds on ρ(x) enable a principled approach to this otherwise unidentified problem. We introduce retrospective counterfactual estimators hatμρ(x,y) and prediction intervals Cρ(x,y) that asymptotically satisfy P[Y(1)∈ Cρ(x,y)mid X=x, Y(0)=y]≥1-α under standard causal assumptions. Many common baselines implicitly correspond to endpoint choices ρ=0 or ρ=1 (ignoring the factual outcome or treating the counterfactual as a shifted factual outcome). Interpolating between these cases through cross-world dependence yields substantial gains in both theory and practice.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 41 of its 67 references that are in the map. First posted on arXiv 28 Mar 2026.
Builds on
41 of 67 references.
References · 41
- 2022TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Secondcited by 643
- 2019Conformalized Quantile Regressioncited by 620
- 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 597
- 2019Conformal Prediction Under Covariate Shiftcited by 511
- 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 475
- 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 408
- 2016Learning Representations for Counterfactual Inferencecited by 327
- 2016Generalized Random Forestscited by 309
- 2017Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processescited by 156
- 2020Conformal Inference of Counterfactuals and Individual Treatment Effectscited by 150
- 2023Conformal Prediction With Conditional Guaranteescited by 146
- 2024Theoretical Foundations of Conformal Predictioncited by 96
29 more
- 2014Consistency of random forestscited by 84
- 2020Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networkscited by 78
- 2017Fast calibrated additive quantile regressioncited by 26
- 2023Conformal Meta-learners for Predictive Inference of Individual Treatment Effectscited by 23
- 2022Bayesian Causal Inference: A Critical Reviewcited by 21
- 2024DiffPO: A causal diffusion model for learning distributions of potential outcomescited by 17
- 2025Rectifying Conformity Scores for Better Conditional Coveragecited by 17
- 2017Causal Inference: A Missing Data Perspectivecited by 14
- 2025Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coveragecited by 12
- 2025Conformal Inference under High-Dimensional Covariate Shifts via Likelihood-Ratio Regularizationcited by 11
- 2025PCS-UQ: Uncertainty Quantification via the Predictability-Computability-Stability Frameworkcited by 9
- 2016Decomposing Treatment Effect Variationcited by 8
- 2020Insights into the "cross-world" independence assumption of causal mediation analysiscited by 7
- 2024Conformal Convolution and Monte Carlo Meta-learners for Predictive Inference of Individual Treatment Effectscited by 7
- 2021On Doubly Robust Inference for Double Machine Learning in Semiparametric Regressioncited by 6
- 2024Counterfactual Cocycles: A Framework for Robust and Coherent Counterfactual Transportscited by 6
- 2024Adapting Prediction Sets to Distribution Shifts Without Labelscited by 6
- 2025CLEAR: Calibrated Learning for Epistemic and Aleatoric Riskcited by 6
- 2016A Framework for Optimal Matching for Causal Inferencecited by 5
- 2023Doubly Robust Counterfactual Classificationcited by 5
- 2023Structural restrictions in local causal discovery: identifying direct causes of a target variablecited by 5
- 2022Nonparametric Estimation of Conditional Incremental Effectscited by 3
- 2025Conformal Prediction Under Generalized Covariate Shift with Posterior Driftcited by 3
- 2024Extreme Treatment Effect: Extrapolating Dose-Response Function Into Extreme Treatment Domaincited by 2
- 2024A Convexified Matching Approach to Imputation and Individualized Inferencecited by 2
- 2025Individual Treatment Effect: Prediction Intervals and Sharp Boundscited by 2
- 2025Asymmetrical Latent Representation for Individual Treatment Effect Modelingcited by 1
- 2025Semiparametric Inference for Counterfactual Regression under Intervention-Driven Shiftcited by 1
- 2025Cross-World Assumption and Refining Prediction Intervals for Individual Treatment Effectscited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.